Real-time packet-switched services, such as voice over Internet Protocol (VoIP), push-to-talk (PTT), and video on demand (VOD), have been widely deployed. Typically, these real-time services perform best when used in a low-latency, low jitter network environment. Accordingly, cellular telephones, personal digital assistants, personal computers, and other types of endpoints may request a particular desired quality of service (QOS) from other network devices in the path of the real-time service. These other network devices, such as switches, routers, firewalls, and access gateways, may then determine whether to honor such a request.
However, new types of real-time applications are emerging that typically do not explicitly request or signal their QOS requirements to other network devices. Instead, these applications, which may be referred to as “third-party” real-time applications, normally rely on network paths between endpoints to have sufficient capacity to provide a reasonable level of QOS. Thus, these applications may rely upon a best-effort level of QOS to provide adequately low latency and jitter.